Monthly Archives: May 2013

Prototyping with Stereolithography

A group of designers built a number of 3D models of a toy robot using Rhino.  Next, they chose to build a reduced scale mock-up of their favorite design.    3-D printing is a great tool for rapid prototype development, because there isn’t the wait for tooling normally associated with other methods.  The 3-D printing process the designers chose was stereolithography.  This is was their model…

Model of Toy Robot

Rhino Model of Toy Robot

In stereolithography, an elevator floor is submersed just below the surface of a vat of a liquid polymer.  This substance has the characteristic that it will harden when exposed to a laser light.  The laser light is precisely focused and aimed to harden the liquid at specific points.  When the floor is just below the surface of the liquid, these hardened points will stick to the floor.  By scanning the beam over the surface of the fluid, a 2-D pattern of hardened points may be arranged on the floor.  After this is done, the floor is lowered slightly into the vat.  The laser scans the surface of the liquid again, attaching another 2-D layer of points onto the last layer.  This process is repeated, again and again, lowering the floor with each iteration.    After the last iteration, a 3-D form, consisting of stacked 2-D layers, can be found immersed in that vat.  The floor is then raised from the vat with the finished product resting atop.

This was the result of the designers’ effort, a 3-D prototype robot with no arms…

Toy Robot Prototype is missing its arms.

Toy Robot Prototype

What went wrong?  When building prototypes with stereolithography, it is sometimes necessary to also design a scaffold to provide a layer for the other layers to “rest upon.”  The designers knew this, but designed a scaffold lacking the necessary rigidity to support the heavy robot arms.  Consequently, the scaffold flexed too far before the arms could be connected to the shoulders.

So, when using stereolithography for your next rapid prototype, remember to include a sufficiently strong scaffold to support the overhanging members.  Otherwise, your rapid prototype might not be prototyped as rapidly as you would have hoped.


copyright 2013, NetChime Research LLC, All Rights Reserved



The Next Big Thing


In 1999, a group of software product developers made a habit of taking morning breaks to get away from their computers and to bounce ideas off of one another.    They would walk over to a nearby coffee shop, share experiences, tell jokes and offer advice.  On average, it was a 20-minute event that occurred three times a week.  Recognizing that this activity was not only building camaraderie, but fostering creative thinking, their manager encouraged them and frequently participated himself.

Of the many topics discussed, two themes kept recurring…  “What’s going to be the Next Big Thing?” and “How are we going to get rich with it?”  Many ideas were suggested, and many were shot down.  After going through this exercise several times, the group came up with the following list identifying characteristics of “The Next Big Thing…”

1)   The idea would have to catch on like an out of control chain reaction.  The first event would lead to other events, and these events would result in more events, and so on.  Think of an avalanche triggered by rolling a single boulder down a hill.

2)   The number of times those events would repeat could not be limited to a small number of potential customers, users or subscribers.    In other words, the market had to be huge.

3)   The person or group triggering the reaction would have to possess the unique capability to do so.  Because most reactive environments aren’t limitless, they didn’t want to start a chain reaction that competitors might notice and then quickly duplicate for their own benefit.  Consequently, the reaction had to happen so quickly that potential competitors could only obtain a negligible market share.   Also, the time necessary to get involved could provide a barrier to delay competition.

4)   The resources necessary to trigger the chain reaction had to be within grasp.
These guys didn’t have lots of resources.  So they needed an idea they could execute without giving away large portions of the opportunity to investors or service providers.

5)   The cost to trigger the first event had to be negligible compared to the expected gains.

6)   The probability of success had to be reasonable.

7)   The plan had to be simple and innovative.

Today, these product developers would probably all agree that having a great idea is only part of the equation.  Along with having a great idea, an excellent plan is required, and the plan must be executed with precision.  The right team is also needed.  This team must have the right skills and a great deal of passion.  Otherwise, the team will not be able to move quickly enough.  This isn’t to say that the list of “next big thing” characteristics was wrong.  Those characteristics are still needed, but those are only the characteristics of the next big idea.  Those aren’t the characteristics of the plan and team that are going to make it happen.

A number of friends from the original group have stayed in touch for years.  Three have started small businesses.   Although one of these businesses was significantly more successful than the others, none of the start-ups hit the spot as far as greatly exceeding the founders’ wildest dreams.   Nonetheless, they did OK.   So, the friends still get together from time to time, and this is what they talk about now…

“What’s going to be the Next Big Thing?” and “How are we going to get rich with it?”


Copyright 2013 NetChime Research LLC,  All rights reserved.

Using the Arduino to Add Analog Inputs to Your Prototype

There is a small company who developed a very cool device with an embedded processor.   To speed their prototype development effort, the company used a commercial-off-the-shelf (COTS) processor board.   This processor came with an open source operating system and some convenient interfaces, which allowed the designers to use the same processor for software development.   They made a nice looking enclosure for the device, found a compatible power supply, and quickly assembled something they would be proud to show a potential investor.

Unfortunately, after all that hard work, they realized they needed an analog input that they didn’t have.  It was for a volume control knob, which they felt would greatly improve the user experience.  At the time they were shopping for the processor board, they didn’t realize they needed that analog input.  Consequently, although being very careful to select something powerful and flexible, they picked something that fell short in this small, but important area.

Here’s how they solved the problem.  They added another processor, one that you don’t see often in commercial prototypes.   They added an Arduino Duemilanove.  Yes, the Arduino is one of the processors you see advertised in electronic hobbyist
e-zines.  Hobbyists love these devices, because they are inexpensive and easy to use.   Well… guess what?  Professionals also love devices that are inexpensive and easy to use.

They could use Arduino, because the original COTS processor had an extra USB interface.  At first, the idea of using an Arduino to provide the analog interface received resistance.   They were told, “You don’t understand the complexities of the USB protocol stack.”  And “This approach will take too long to code up. “

So they took a little time to investigate the approach.  This is what they found.  It was easy, so easy, in fact, that they were able to demonstrate the solution in less than 4 hours.   They connected a little potentiometer up to the Arduino with three wires.  That was that easy.  They found a pre-written script for sampling an analog input and transmitting those samples over the Arduino’s USB interface.   And that was easy.  They downloaded and installed the Arduino development software and then loaded the Arduino sampling script into the Arduino.  Next, they connected the Arduino to the COTS processor.  They had to write a little shell script to run on the open source operating system to see their results, but that was easy too.    This was just another serial device as far as the operating system was concerned.

What was really nice was that the Arduino got its power through the USB interface.  So, it was not necessary to provide another voltage source to power the Arduino.

Also, since the Arduino is small, they could fit it inside the existing enclosure.

In the field of rapid prototype development, you need to be creative, and you need to think out of the box.  But most of all, you need to be quick.   If you’re taking too long to implement the next wiz bang device, you might just be allowing your competition to beat you to the market.  That can mean missing a grand opportunity.  The company realized this, and they did what needed to be done.

After the accomplishment, the team jokingly claimed “we have just constructed the world’s most expensive volume control knob,” but they all knew the truth.   When it comes to prototype development, time is money, and this was the least expensive way to achieve the desired result and still make their schedule.

copyright 2013 NetChime Research, LLC,  All rights reserved.

Holistic Security

Once upon a time, back around 1998, there was this network security specialist.  He won a small contract to evaluate the network security stance of a building owned by a large organization that was “serious” about security.  Eager to provide excellent service, he went to work in earnest.  He was quickly rewarded when he found the organization did indeed have some vulnerabilities, most notably, a situation, where, through an SQL injection attack, the organization’s entire database was completely exposed.  This will impress my customer, he thought, but he dug deeper.  Eventually, he found evidence that the company had already been hacked.  He documented this for his “final report,” but kept digging.  He then found that this wasn’t the first time this organization had been hacked.  It had happened more than once before.  In fact, this network was the scene of a battle between two groups of hackers that had been fighting against one another for months.   The two groups had been taunting each other, which simplified the job of tracking them.   Surely, this fact would impress the customer, but the network security specialist went further and found evidence of another intruder, one who was a bit more cunning, one who was far more sophisticated in covering his tracks.  It was fortuitous that this intruder couldn’t resist protecting his captured territory by removing the vulnerabilities that might allow others to follow him.  This fact was crucial in his detection.

Eventually, the money ran out, the network security specialist stopped digging, the final report was written, and the contract ended.    The network security specialist was a disappointed.  There was obviously more work to be done.  What bothered him most was that he knew the network under examination was connected to other networks and other “friendly” organizations that probably had similar problems.   One of recommendations in the final report was “These other networks should also be examined…”  Certainly, if intruders can get into one of these networks, intruders would have no problem getting back into the network under examination.

Not knowing the story from another perspective, one may wonder why the network security specialist wasn’t rewarded with more work.   Perhaps it was a matter of money.  The IT manager that hired him obviously didn’t have an unlimited supply, and specialists do cost money.  Specialists find problems that need to be fixed.  This takes time too.   It’s also possible the network security specialist had unwittingly endangered the reputation of his customer.  His work product was a report that, if mishandled, could make the customer look negligent.  After all, in the final analysis, it was the IT manager’s responsibility to make sure things like this didn’t happen.  Imagine his feelings on learning about what had been happening right under his nose.

In the end, the decision of whether to keep the network security specialist was probably made based on the principals of Holistic Security.  The IT manager had something he valued, an asset like his budget, his time or his reputation.  The network security specialist represented a threat to that asset, and a decision was made not to renew the contract.   This brings us to an interesting observation…

Everyone, whether they know it or not, is a security specialist.    Everyone has something they value, and everyone will do things to enhance and protect those things.

Holistic security is about being continuously aware of

  1. Assets: The things we have and value
  2. Threats:  The things that threaten our assets
  3. Security Functions:  The things that protect us from these threats
  4. Opportunities:  The things we don’t have and want
  5. Barriers:  The things that prevent us from seizing these opportunities

It is important that this awareness is continuous, because these things can be changing dynamically.  More than awareness, Holistic security is also about taking the action to improve the situation, when possible, in a balanced way.  Accordingly, obtaining some valued thing is sometimes not worth risking the valued things you already have.  To be excellent at Holistic Security, one must be ever vigilant, paying special attention to game changing events and trends.  So, we are all in the game, and we all play to win.  Unfortunately, not everyone will win.  This goes for the network security specialist too.

copyright 2013 NetChime Research LLC, All rights reserved